Solving Inventory Management Problems

Henry C. Co Technology and Operations Management, California Polytechnic and State University

Inventory
Inventory: stockpiles of raw materials, components, semi-finished or finished goods waiting to be processed, transported or used at a point of the supply chain. Reasons to have inventories
l l l l l

Improving service level, Reducing overall logistics costs, Coping with randomness in demand and lead times Making seasonal items available all year Speculating on price patterns, etc.

Annual inventory holding cost can be 30% of the value of the materials kept in stock, or even more.
Inventory Management Problems

Ghiani, p. 121
2

Relevant Cost
Procurement costs Holding costs Shortage costs Obsolescence costs

Ghiani, p. 122
Inventory Management Problems 3

Inventory Management Models

Deterministic vs. stochastic models Fast- vs. slow-moving items No. of stocking points No. of commodities Instantaneous resupply v. non-instantaneous resupply Discrete vs. continuous; finite vs. infinite horizon; shortage allowed/disallowed etc.
The Economic Order Quantity model you learned in TOM 301 is an example of a single stocking point, single-commodity, instantaneous resupply, shortage not allowed, continuous model with infinite horizon.
Ghiani, p. 123
Inventory Management Problems 4

The EOQ Model in TOM 301

The total cost curve reaches its minimum where the carrying and ordering costs are equal. EOQ represents trade-off between fixed cost associated with production or procurement against inventory holding costs.
D= S= v = H= Q= Rate of demand, units/year Fixed cost of procurement, $/order Variable cost of procurement. $/unit/year holding cost Quantity ordered, units

Inventory Management Problems

Instantaneous Resupply

Ghiani, p. 129
Inventory Management Problems 6

The Classical EOQ Model

The total cost curve reaches its minimum where the carrying and ordering costs are equal.
Annual Cost

TC =

Q
2

D S Q

Ordering Costs (optimal order quantity) QO

Inventory Management Problems 7

Order Quantity (Q)

Using calculus, we take the derivative of the total cost function (TC) and set the derivative (slope) equal to zero and solve for Q.
Q OPT = 2DS = H 2(Annual Demand)(Or der or Setup Cost) Annual Holding Cost

Inventory Management Problems

Try This!
Al-Bufeira Motors manufactures spare parts for aircraft engines in Saudi Arabia. Its component Y02PN, produced in a plant located in Jiddah, has a demand of 220 units per year and a unit production cost of $1200. Manufacturing this product requires a time-consuming set-up that costs $800. The current annual interest rate p is 18%, including warehousing costs. Shortages are not allowed.

Ghiani, p. 123
Inventory Management Problems 9

Non-instantaneous Resupply (Batch Production Model)

Ghiani, p. 126

Suppose production rate = p= 200 units/day, and demand rate = d = 80 units/day. Since p>d, inventory will increase at __________ (p-d) units/day. Suppose current inventory is 0.
l In

10 days, the inventory level would be 10 days * 120 unit/day = _____ units. l In 20 days, the inventory level would be 20 days * 60 unit/day = _____ units. l etc.

Inventory Management Problems

11

The machine produces a batch, then stops, then resumes production at some later time when the inventory of this item is low. This is call batch production. Batch production is very common in industry.
l l

When a machine is used to produce two or more products, one product at a time. One decision the production manager has to make is when to start producing each product, and when to stop.

The run time is the amount of time the machine is producing a batch.
l

Producing at 200 units/day, if we want to produce 2,000 units per batch, the run time is _____ days.

Inventory Management Problems

12

Maximum Inventory Level

If current inventory level is 0, what is the inventory at the end of the run time?
l

Since inventory will be rising at (200- 80 =120) units/day, the inventory level will be _____ units in 10 days. The inventory at the end of the run time is the maximum inventory. It is equal to (p-d)*t = _____ units.

The machine produced 2,000 units in 10 days, and the maximum inventory level is only 1,200. Why? After completing a batch, how long will it take to deplete the inventory?
l

Answer: It will take (p-d)*t/d = _____ days to deplete the inventory. This is the off-time.

Inventory Management Problems

13

Number of Runs Per Year

If annual = D = 24,000 units, how many runs do we produce each year?
l Answer:

Since we are producing Q = 2,000 units per batch, there will be D/Q = _____ batches per year. l In other words, there are D/Q = 12 cycles per year. In each cycle, there is a period of time the machine is producing the product (the run time), and a period to allow the inventory to deplete (the off time).

Inventory Management Problems

14

Average Inventory
What is the average inventory level?
l

During the run time, the inventory level rises from 0 to the maximum level of (p-d)*t = _____ units. During the off time, the inventory level drops from a maximum of (p-d)*t units to 0. The average inventory level therefore = _____ units.

Since p*t = Q, then t = Q/p. We can rewrite the expression for the average inventory as (p-d)*t/2 = (p-d)*(Q/p)/2 = (1-d/p)Q/2.

Inventory Management Problems

15

Tradeoff
What is the average inventory if the batch size equals the annual demand D = 24,000 units? How many batches do we have to run per year?
l

Answer: The average inventory = (1-d/p)Q/2 = _____ units. We need to run one batch per year.

What is the average inventory if the batch size equals the weekly demand of 480 units (assuming 50 weeks/year)? How many batches do we have to run per year?
l

Answer: _____ units; Run _____ batches per year.

Inventory Management Problems

16

Optimal Tradeoff
Suppose the cost to carry one unit of inventory for one year is H. Since the average inventory level is (1-d/p)Q/2, the annual inventory-carrying cost is H*(1-d/p) Q/2. Suppose the cost to set-up the machine to produce a batch is S. Since we need to run D/Q batches per year, the annual set-up cost is S*D/Q. Adding the two costs, we have H*(1-d/p) Q/2 + S*D/Q. Using calculus, the optimal batch size is

Inventory Management Problems

17

Try This!
Golden Food distributes tinned foodstuff in Great Britain. In a warehouse located in Birmingham, the demand rate d for tomato pure is 400 pallets a month. The value of a pallet is c = 2500 and the annual interest rate p is 14.5% (including warehousing costs). Issuing an order costs 30. The replenishment rate r is 40 pallets per day. Shortages are not allowed.

Ghiani, p. 126
Inventory Management Problems 18

Quantity Discounts-On-All-Units

Ghiani, p. 133
Inventory Management Problems 19

Try this!
Maliban runs more than 200 stationery outlets in Spain. The firm buys its products from a restricted number of suppliers and stores them in a warehouse located near Sevilla. Maliban expects to sell 3000 boxes of the Prince Arthur pen during the next year. The current annual interest rate p is 30%. Placing an order costs 50. The supplier offers a box at 3, if the amount bought is less than 500 boxes. The price is reduced by 1% if 500 2000 boxes are ordered. Finally, if more than 2000 boxes are ordered, an additional 0.5% discount is applied.

Ghiani, p. 134
Inventory Management Problems 20

Single Period Stochastic Models

Short product life cycles / Long lead times

l l

Computers Apparel Fresh food, newspapers

Fresh products
l

Services Airline industry

These models have the objective of properly balancing the cost of Underage having not ordered enough products vs. Overage having ordered more than we can sell These models apply to problems like:
l l l

Planning initial shipments of High-Fashion items Amount of perishable food products Item with short shelf life (like the daily newspaper)

Because of this last problem type, this class of problems is typically called the Newsboy problem
Inventory Management Problems 22

Stochastic Model 1: The Newsboy Model

At the start of each day, a newsboy must decide on the number of papers to purchase. Daily sales cannot be predicted exactly, and are represented by the random variable, D. The newsboy must carefully consider these costs:
l

cU: underage cost (when DS). This is the unit opportunity cost; for example, unit revenue r unit cost c, i.e., (r-c) cO: overage cost (when DS). This is the unit cost of overstocking; for example, unit cost c - unit salvage value u, i.e., (c-u).

Inventory Management Problems

23

The objective is to Minimize the expected cost: cu E[max{D-S, 0}] + co E[max{S-D, 0}] Solving for Q, the optimal order quantity S satisfies the following condition:

Equation (4-36) can be rewritten as:

cu r c r c r c = = = r u r u + (c c) (r c )(c u ) cu + c0
Ghiani, p. 141-142
Inventory Management Problems 24

Graphical Representation

Inventory Management Problems

25

Uniform Demand Between [A,B]

Solve

cu S* A = B A cu + co

S = A+ (B A)
*

cu

cu + co r c = A+ (B A) r u
26

Inventory Management Problems

Try this on Excel!

Emilio Tadini & Sons is a hand-made shirt retailer, located in Rome (Italy), close to Piazza di Spagna. This year Mr. Tadini faces the problem of ordering a new bright color shirt made by a Florentine firm.
l l

He assumes that the demand is uniformly distributed between 200 and 350 units. The purchasing cost is c = 18 while the selling price is r = 52 and the salvage value is u = 7. Thus co = 34 and cu = 11.

S * = A+ (B A)

cu

cu + co 52 18 r c = A+ (B A) = 200 + (350 200 ) = 313 52 7 r u

= 200 + (350 200 )

34 = 313 34 + 11

Hence, Mr. Tadini should order S = 313 units.

Ghiani, p. 142
Inventory Management Problems 27

Download Excel worksheet

Inventory Management Problems 28

Another Example
The buyer for Needless Markup, a famous high end department store, must decide on the quantity of a high-priced womens handbag to procure in Italy for the following Christmas season.
l l l

The unit cost of the handbag to the store is $28.50 and the handbag will sell for $150.00. Any handbags not sold by the end of the season are purchased by a discount firm for $20.00. In addition, the store accountants estimate that there is a cost of $.40 for each dollar tied up in inventory, as this dollar invested elsewhere could have yielded a gross profit. Assume that this cost is attached to unsold bags only. Suppose that the sales of the bags are equally likely to be anywhere from 50 to 250 handbags during this season.

Based on this, how many bags should the buyer purchase?

Example from Nahmias, Production and Operations Analysis
Inventory Management Problems 29

Equation (4-36) does not account for inventory-holding cost!

Inventory Management Problems 30

Distribution of Demand Is Normal

Inventory Management Problems

31

Example
Every week, the owner of a newsstand purchases a number of copies of The Computer Journal.
Weekly demand for the Journal is normally distributed with mean 10 and standard deviation 5. l He pays 25 cents for each copy and sells each for 75 cents.
l

Question: How many copies should he order?

Example from Nahmias, Production and Operations Analysis
Inventory Management Problems 32

Using NORMSINV

Inventory Management Problems

33

Using NORMINV

Inventory Management Problems

34

Using Table

Inventory Management Problems

35

Stochastic Model 2
(s, S) Policy for Single Period :
l

If there is an initial inventory q0 and a fixed reorder cost k, the optimal replenishment policy can be obtained as follows. If q0 S, no reorder is needed. Otherwise, the best policy is to order S q0, provided that the expected revenue associated with this choice is greater than the expected revenue associated with not producing anything.

Hence, two cases can occur:

(i) if the expected revenue (S) k cq0 associated with reordering is greater than the expected revenue (q0) cq0 associated with not reordering, then S q0 units have to be reordered; (ii) otherwise, no order has to be placed.
Ghiani, p. 142
Inventory Management Problems 36

Stochastic Model 3
In reorder point policy (fixed order quantity), inventory level monitored continuously. As soon as its net value I(t) (amount in stock - unsatisfied demand + orders placed but not yet received) reaches a reorder point l, a constant quantity q is ordered.

Ghiani, p. 143
Inventory Management Problems 37

Stochastic Model 4
In the reorder cycle policy (periodic review policy) stock level is reviewed periodically at time instants ti (ti+1 = ti + T , T0). At time ti, qi = S I (ti) units are ordered. The order-up-to-level S represents the maximum inventory level in case lead time tl is negligible.

Ghiani, p. 145
Inventory Management Problems 38

Stochastic Model 5
The (s, S) inventory policy is a natural extension of the (s, S) policy illustrated for the one-shot case. At time ti , S I(ti) items are ordered if I(ti) < s. If s is large enough (s S), the (s, S) policy is similar to the reorder cycle inventory method. On the other hand, if s is small (s 0), the (s, S) policy is similar to a reorder level policy with a reorder point equal to s and a reorder quantity q S.

Ghiani, p. 147
Inventory Management Problems 39

Stochastic Model 6
The two-bin policy is a variant of the reorder point inventory method where no demand forecast is needed, and the inventory level does not have to be monitored continuously. The items in stock are assumed to be stored in two identical bins. As soon as one of the two becomes empty, an order is issued for an amount equal to the bin capacity.
Browns supermarkets make use of the two-bin policy for tomato juice bottles. The capacity of each bin is 400 boxes, containing 12 bottles each. In a supermarket close to Los Alamos (New Mexico, USA) the inventory level on 1 December last was 780 boxes of 12 bottles each. Last 6 December, the inventory level was less than 400 boxes and an order of 400 boxes was issued (see Table 4.2). The order was fulfilled the subsequent day.

Ghiani, p. 147-148
Inventory Management Problems 40

Simulation of the (S, S) Policy Excel Worksheet

The (s, S) policy is a good compromise between the reorder level and the reorder cycle policies. Unfortunately, parameters T , S and s are difficult to determine analytically. Therefore, simulation is often used in practice.

Inventory Management Problems

41

Homework
The Ortiz County Hospital Blood Bank Case